Re: UK Project Opportunity

[email protected] (Nathan Rixham)
Newsgroups php.general
Message-ID <[email protected]>
Manuel Lemos wrote:
> Hello,
> 
> on 02/17/2010 12:34 AM Nathan Rixham said the following:
>>>> in all honesty the following are the sticking point that make it hard
>>>> for me to decide if I was right in my earlier response:
>>>>
>>>> 1: that the "sign up to get the class" is even part of the equation
>>> This is another misunderstanding. The site does not make anybody
>>> register to download any package. That is an option determined by each
>>> author.
>>>
>>> That is explained in the site FAQ and other places, but some people
>>> still misunderstand it and assume that it is an evil imposition of the site.
>>>
>>> http://www.phpclasses.org/faq/#register-to-download
>>>
>>> It may not be obvious, but the truth is that this detail is one of the
>>> reasons for the success of the site.
>> again, hence why I said "part of the equation"; it is can be switched on
>> and off by the author on a per file basis, but to me it would make more
>> sense to leave the decision of signing up to the user, and not try and
>> influence it in anyway - personally I would remove the optional part of
>> it and simply do what other sites do; present the user with two choices
>> "register and download" or "download without registering".
>>
>> A similar topic of opt-in vs opt-out on emails is often discussed;
>> however I believe it has to be "opt-in" now, within the UK certainly.
> 
> That is because you are only seeing it from the view point of the user.
> 
> People need to have motivation to do what they do. Authors only submit
> their packages to PHPClasses site if they gain something from it. What
> the site provides them is attention.

I know - I'm an author ;) [or was, well still am, can't delete profile
so left some files up their for anybody who may find them useful] - and
as for attention, the last class I put on phpclasses was getting 5
downloads a week, and 1k+ a week on sourceforge.

> As I explained in another reply, if an author leaves the login
> requirement enabled, the site can keep track of the users that
> downloaded the packages. That can be used for sending alerts when a new
> version of the package is released, or a new blog about the package is
> posted.
> 
> The site can also build download charts that lets authors see their
> progress in user interest and motivate them to continue to submit more
> packages or update existing ones.
> 
> In the end everybody wins.

and it can do all this with the user being in control and choosing to
give that info.. as implemented on millions of other sites..

>>>> 3: that users can't delete an account
>>> That is explained in the FAQ. The site does not allow users to recreate
>>> accounts with the same address or access name. If you really delete an
>>> account, you have no way to prevent that an user recreates an account
>>> with the same access name and e-mail address. That would be a hole for
>>> malicious users to cheat on several types of rankings and contests that
>>> the site organizes.
>>>
>>> http://www.phpclasses.org/faq/#delete-account
>>>
>>> Other than that, most sites out there never really delete accounts. Some
>>> claim they do, but then you try to create an account with the same
>>> credentials and the site says the account still exists. I would rather
>>> tell users the truth.
>> I've read it quite a few times, and as a developer I know very well that
>> you could delete the account and prevent the address from being re-used
>> very easily with little code; the ethics of making somebody remove each
>> tiny bot of info and file from your site in order to manually remove
>> themselves is a bit.. debatable to me - again, personally I'd do it the
>> same way as everybody else, giving the user the option of whether to
>> make the username and email address available to be re-registered. I
>> strongly feel all these kind of choices should be in the hands of the
>> user, and not the company.
> 
> The site sorts of give that option to the users by letting them know
> that it does not remove accounts. Users that disagree, should just not
> register.
> 
> It is like you go to some country and if for some reason you leave,
> probably upset with something, you will not be able to erase the records
> of your presence in  that country.
> 
> This site is not really interested in your personal information. The
> reason this site does not remove accounts is because it is used to
> prevent several types of fraud. I am just not going to enter in much
> detail here in public on what happens to not give ideas to cheaters that
> may be reading this. I hope you understand.

as stated, you could just remove the public pages of said user and keep
the username + email taken; quite easily

>>>> 4: the amount of adverts
>>> I wish it was viable to remove all the ads. That would mean that the
>>> site could succeed on revenue from other services. Advertising was plan
>>> B. I still hope someday I can take all ads down. Meanwhile, premium
>>> subscribers have all ads removed from the site. In some cases, they see
>>> other valuable information in the place of the ads.
>> the amount; not suggesting you remove adverts all together, it's the
>> currency of the web!
> 
> Unfortunately I cannot reduce the current amount of ads. The site
> already eliminated most types of pop-under ads because they are too
> annoying for the users. That caused a significant ad revenue reduction.
> 
> Maybe if the number of premium subscriptions raises to a significant
> level I can reduce or even eliminated all ads. While that does not
> happen, I am afraid the current number of ads has to kept.

that's a business decision then I guess.

>>> There are much more things upcoming. Most of the things are based on
>>> suggestions from users because that is the way to make this a more
>>> satisfactory project for everybody.
>>>
>>> But I have to manage this as any real business to keep it viable.
>>>
>>> Most people that use the site do it to some how make money for
>>> themselves, being that using code or solutions found on the site to
>>> develop paid projects for clients, find developers to work on their
>>> projects, etc..
>>>
>>> The site charges for some services, but I believe it charges reasonable
>>> amounts for the benefits it provides and the paying users will make much
>>> more using such benefits.
>>>
>> I do hope it all goes well for you and the community, and that the right
>> choices are made - when in doublt just look at the big guys like
>> sourceforge, github etc and see how they do it - the better your service
>> and easier it is for users, the more chance you get of full donated
>> hosting and investments - I guess a good measure will be when you have
>> several of the big PHP projects / libraries on there.
> 
> Well, those sites are not exactly for the same purpose. They are for
> hosting the actual project development. The PHPClasses site is more for
> distribution. Each type of site needs to find a business model that
> adequates to the type of activity that goes one in the site.
> 

if you made the site more accessible; didn't lock in users, took off the
registration, cut down the ads (not remove), improved the templates,
promoted the developers by giving their details free and showing off
their fine work; hell why not even integrate with these fine project
hosting sites (that'll cut the bandwidth..) then maybe - just maybe
traffic would rise; people would enjoy the site more; use it more and in
turn the money maker which is the job ad's would get more exposure and
boost revenue for you.

you've got everything there to make a great community which would in
turn fund you and be a fine business - there are other ways to do
things. I know it's been 8 years but you can always change / tweak.

ultimately though, it is a business the focus has to be on making sure
the money comes in & that has to be priority #1 (?)
lmpx.com only provides a reader for public news (NNTP) servers. It is not affiliated with the servers or forums shown here and is not responsible for the content of articles, which is written by their respective authors.